Relationship between antimycobacterial activities of rifampicin, rifabutinand KRM-1648 and rpoB mutations of Mycobacterium tuberculosis

We compared the in-vitro antimycobacterial activities of rifabutin and KRM- 1648, two rifamycin derivatives, with that of rifampicin against 163 strain s of Mycobacterium tuberculosis. We also evaluated the correlation between the level of resistance to rifampicin, rifabutin and KRM-1648 and genetic a lterations in the rpoB gene. All 82 strains susceptible to rifampicin or re sistant to rifampicin with MICs less than or equal to 16 mg/L were suscepti ble to rifabutin and KRM-1648 with MICs less than or equal to 1 mg/L. Seven ty-six of 81 strains resistant to rifampicin with MICs greater than or equa l to 32 mg/L were resistant to both rifabutin and KRM-1648, but with lower MICs than those of rifampicin. KRM-1648 showed more potent antimycobacteria l activity than rifabutin against organisms with low MICs (less than or equ al to 1 mg/L), while rifabutin was more active than KRM-1648 against organi sms with high MICs (greater than or equal to 2 mg/L). A total of 96 genetic alterations around the 69 bp core region of the rpoB gene were detected in 92 strains. Alterations at codons 515, 521 and 533 in the rpoB gene did no t influence the susceptibility to rifampicin, rifabutin and KRM-1648. Point mutations at codons 516 and 529, deletion at codon 518 and insertion at co don 514 influenced the susceptibility to rifampicin but not that to rifabut in or KRM-1648. With the exception of one strain, all alterations at codon 513 and 531 correlated with resistance to the three test drugs. The resista nt phenotype of strains with an alteration at codon 526 depended on the typ e of amino acid substitution. Our results suggest that analysis of genetic alterations in the rpoB gene might be useful not only for predicting rifamp icin susceptibility, but also for deciding when to use rifabutin for treati ng tuberculosis. Further studies may be required to determine the usefulnes s of KRM-1648.